CD4+CD25highCD127low/−FoxP3+ regulatory T cells (Tregs) are currently under extensive investigation in childhood acute lymphoblastic leukemia (ALL) and in other human cancers. Usually, Treg cells maintain the immune cell homeostasis. This small subset of T cells has been, in fact, considered to be involved in the pathogenesis of autoimmune diseases and progression of acute and chronic leukemias. However, whether Treg dysregulation in CLL and ALL plays a key role or it rather represents a simple epiphenomenon is still a matter of debate. Treg cells have been proposed as a prognostic indicator of the clinical course of the disease and might also be used for targeted immune therapy. Our study revealed statistically higher percentage of Treg cells in the bone marrow than in peripheral blood in the group of 42 children with acute lymphoblastic leukemia. By analyzing Treg subpopulations, it was shown that only memory Tregs in contact with leukemic antigens showed statistically significant differences. We noticed a low negative correlation between Treg cells in the bone marrow and the percentage of blasts (
Acute lymphoblastic leukemia (ALL), the most common childhood cancer, is a heterogeneous disease that occurs due to the malignant clonal proliferation of lymphoid progenitors [
Normal cells in the environment of cancer cells are currently under intensive investigation. Residual nonmalignant T cells and B cells are in permanent cell-to-cell contact with lymphoblasts and are involved in active immune responses [
Regulatory lymphocytes constitute a very interesting subpopulation of cells of the human immune system. A growing interest in their biological properties has occurred recently and clinicians have wondered whether they can be also used in the battle against cancer [
Recent papers have demonstrated elevated number of Tregs in lung, breast, pancreatic, ovarian, melanoma, digestive system cancers, CLL, T cell ALL, and B cell NHL [
It was demonstrated previously that elevated percentages and increased suppressor properties of Treg cells are observed even after achieving a remission and after completing the treatment of AML [
Previous research showed that the number of Treg cells may be either elevated or reduced in Hodgkin disease and mature B cell lymphoma. Similarly, a prognosis may be either favorable or adverse [
In our study, we investigated a population of CD4+CD25highCD127low/–FoxP3+ regulatory T cells in the bone marrow and peripheral blood of children with acute lymphoblastic leukemia treated in the Department of Pediatrics, Hematology and Oncology, Medical University of Gdansk in 2011–2016.
Due to the small number of publications concerning the influence of Tregs on the prognosis in acute childhood leukemias and investigating the percentage of these cells in the bone marrow and peripheral blood of ALL children, a following research was undertaken to understand these relationships better.
Of particular interest to us was the influence of a higher percentage of Tregs in the peripheral blood/bone marrow of patients with acute leukemia on the early and late therapeutic effect, which was reported in the literature [
In addition, it was decided to perform an initial assessment of the relationship between biological characteristics of leukemia and Tregs. By assessing the correlation between the number of Tregs and such parameters as hemoglobin, platelets, leukocytosis, or the percentage of blasts in the peripheral blood and bone marrow at the moment of diagnosis, it was decided to verify the preliminary hypothesis which assumes the connections between Tregs and the stage of the cancer process and prognosis. The elevated percentage of Treg cells in the bone marrow observed earlier by some authors in comparison to peripheral blood requires verification due to the use of a narrow panel of antibodies to assess the population of cells of interest. To effectively and reliably count Treg cells in the analyzed material, antibodies were used to identify cells with the CD4+CD25highCD127low/−FoxP3 phenotype, which identifies the T-line regulatory cells in the most accurate way [
In case the relationships described earlier in the literature were confirmed, it would be quite advisable to search for therapeutic methods interfering with the immune system through manipulations on Treg cells [
An interesting question was also whether the increased percentage of Tregs in the peripheral blood and/or bone marrow is also observed in children with cancers other than leukemia [
In summary, Tregs are a potential target of immunotherapy but this hypothesis requires further, intensive investigation of the properties of relationships between regulatory and cancer cells. This could contribute to the improvement of a prognosis with simultaneous reduction of toxic chemotherapy [
The bone marrow and peripheral blood were obtained at diagnosis from 42 patients with acute lymphoblastic leukemia treated according to the BFM SG Protocol ALL-IC BFM 2002 (
All clinical data concerning patients are summarized in Table
Patient characteristics (
Age | 1–5 years = 26 | 6–11 years = 10 | 12–18 years = 6 |
---|---|---|---|
Gender | Male = 21 | Female = 21 | |
Immunophenotype | B cell = 40 | T cell = 2 | |
Protocol of therapy | ALL IC BFM 2002: 1 | ALL IC BFM 2009: 41 | |
Risk group | SR: 5 | IR: 25 | HR: 12 |
CNS involvement | Positive: 2 | Negative: 40 | |
EFS | Relapse: 2 | Death: 2 (after relapse: 0) | Live in first remission: 38 |
Steroid sensitivity |
Good: 35 | Poor: 6 | |
BM on day 15 |
M1: 28 | M2: 9 | M3: 4 |
BM on day 33 |
M1: 38 | M2: 1 | M3: 2 |
This study was approved by the Medical University of Gdansk Ethical Board, and informed consent was obtained from patients and/or their legal guardians.
A response to the steroid therapy was checked in the peripheral blood on day 8, a remission in the bone marrow was checked by the flow cytometry on day 15 and 33. Patients were divided to SR, IR, and HR risk groups according to the protocol rules.
In the control group, 10 bone marrow samples (2 ml) and 46 peripheral blood samples (5 ml) were tested. For ethical reasons, the bone marrow was obtained only from the children requiring a diagnostic bone marrow biopsy to exclude bone marrow involvement by a cancerous disease or to exclude leukemia. In the control group, the following diagnoses were noted: Wilms tumor (
The analysis involved 24 girls and 22 boys from 1 to 16 years of age.
Only freshly obtained samples were processed up to 24 hours from collection. Briefly, lymphocytes were isolated using density gradient media Lymphoprep (STEMCELL Technologies, Canada) and EDTA bone marrow or peripheral blood samples. Lymphocytes were then stained with the use of CD127 FITC (clone HL-7R-M21), CD25 PE (clone 2A3), CD4 PerCP (clone SK3), CD3 V450 (clone UCHT1), CD45RA PEcy7 (clone L48), and CD62L Alexa Fluor750 (clone Dreg-56). All of the antibodies were obtained from BD Bioscience, USA, except for CD62L from Life Technologies, USA. Permeabilization was done with the use of Foxp3 Staining Buffer Set Kit (eBioscience, USA), while for intracellular staining, FoxP3 APC was used (clone PCH 101, eBioscience, USA). The readout was done with BD FACSCANTO II (BD Bioscience, USA) and 100.000 of cells were acquired.
A representative example of Treg subpopulations gating is given in Figure
First, singlets were identified according to FSC area to height signal distribution (A). Then lymphocytes (B) and CD3+/CD4+ T lymphocytes were gated (C). Next, regulatory T cells were identified as CD4+/FoxP3 double-positive T cells (D), as well as CD127low/CD25+ T cells (E). To get the best overlay between CD127low/CD25+ and FoxP3, Treg gate was put to get minimum 90% of cells in that were FoxP3 positive (F). Then another gate was plotted to identify naïve Tregs as CD45RA+/CD62L+ and memory Tregs as CD45RA–/CD62+ T lymphocytes (G).
Clinical data, laboratory findings, and family history of the disease were collected in the medical database constructed in Microsoft Excel software for Windows 10 (Microsoft). Data were analyzed using Statistica software version 7.1 for Windows (StatSoft Inc. 2005). Shapiro-Wilk test was used to estimate either normal or abnormal spread of analyzed variables. Depending on the spread of variable, nonparametric Mann–Whitney
In our study, for the first time, the percentage of individual subpopulations of regulatory T cells (Tregs) among CD3+CD4+ lymphocytes in the bone marrow and peripheral blood of children suffering from acute lymphoblastic leukemia were determined (Table
Distribution of tested parameters in the bone marrow and peripheral blood in children at diagnosis of ALL.
Tested parameters | Average (%) | Min-max (%) | (Event count) min-max | SD |
---|---|---|---|---|
Treg BM | 9.59 | 2.23–19.03 | 3528–9913 | 3.58 |
Treg PB | 7.81 | 3.33–13.36 | 2183–6942 | 2.73 |
Natural Treg BM | 5.39 | 1.00–16.40 | 1574–7391 | 3.80 |
Natural Treg PB | 3.85 | 0.70–12.00 | 867–5855 | 2.38 |
Natural naive Treg BM | 3.86 | 0.10–8.20 | 569–4003 | 1.98 |
Natural naive Treg PB | 3.80 | 0.90–9.50 | 1161–5814 | 2.14 |
Bone marrow and peripheral blood percentage of Tregs in CD4+ population of cells among children with ALL.
Percentages of regulatory T cells were significantly higher in the bone marrow (9.59+\–3.58) as compared to the peripheral blood (7.81+\–2.73) (
When the percentage of Tregs in the bone marrow was measured, the analysis showed no statistically relevant differences between children suffering from ALL and those diagnosed with solid tumor/anemias/lymphadenopathy or even healthy ones. But when the peripheral blood was taken under investigation, there was statistically higher percentage of Tregs among children in the control group in comparison to pediatric patients diagnosed with ALL (Figure
Treg level in peripheral blood among children with ALL versus control group.
Memory Treg proportion in the bone marrow of children with ALL was statistically higher than the percentage in peripheral blood (
Statistical analysis of selected Treg subpopulations in analyzed population of ALL children.
To identify prognostically relevant parameters, we analyzed correlation between Treg subpopulations and ALL well-known risk factors and a response to treatment. The risk factors and disease parameters analyzed in our study were gender, age, leukocytosis, and blastosis in the peripheral blood, bone marrow blast count at diagnosis, CNS status (M1; M2; M3 according to Protocol ALLIC 2009), and initial qualification to the risk group (SR, IR, and HR).
The response parameters analyzed in our study were blasts’ sensitivity for prednizon (number of blasts on the 8th day of therapy), the percentage of blasts in the bone marrow on day 15 and 33 (<5% (M1 status) or 5% to 20% (M2) or >20% (M3)), and white blood cell count in peripheral blood on day 15 and 33.
Statistical analysis revealed a few interesting observations. A low negative correlation was noticed between Treg cells in the bone marrow and the percentage of blasts in peripheral blood (Figure
Correlation between the percentage of Tregs in BM and percentage of blasts in PB at the moment of diagnosis (
Correlation between the percentage of Tregs in BM at the time of diagnosis with the hemoglobin level (
Correlation between blasts level on the 8th day of steroid therapy in PB with the percentage of Tregs in BM at the moment of diagnosis (
Statistical analysis revealed a low negative correlation between the level of Tregs in the BM and the minimal residual disease measured on day 15 (MRD 15;
ALL is one of the most common childhood cancer with favorable prognosis. Less than 20% of children with acute lymphoblastic leukemia have unfavourable prognosis and suffer from a relapse, resistant ALL, or serious complication of the chemotherapy [
Treg cells play a key role in human immunological reaction towards the neoplasmatic cells in the organism. An increased number of Treg cells was noticed in many solid tumors, for example, breast, colon, and lung tumors [
Generally in human cancers, higher percentages of Tregs predict worse immunological reaction to the viral infection and cancer antigens. However, the role of Treg cells in the pathogenesis of ALL and AML is still unclear [
Tregs are defined on the basis of combined expression of CD4, CD25, FoxP3, low expression of the CD127, and CD4+CD25highCD127low/–FoxP3+ regulatory T cell phenotype is the most appropriate one. This sensitive and reliable phenotype was used to determine the percentage of these cells in the bone marrow and/or peripheral blood of children with ALL.
So far, no data has been reported on Treg cell number in the bone marrow among children with acute leukemias. Similarly, very limited evidence is reported about Treg cells in children leukemias and the influence of their number on the prognosis and their correlation with already known risk factors. Lustfeld et al. suggest that elevated proportions of CD4+ T cells among residual bone marrow T cells in ALL is associated with favorable early responses [
Our analyses confirm these observations and indirectly indicate a correlation between the percentage of Tregs and prognosis in pediatric ALL. The correlations between all risk factors and hematological parameters of ALL patients and peripheral blood and bone marrow Treg number were analyzed. Probably due to the small size of the group, it was possible to detect only the low negative correlation (
Thus, the correlation between blasts’ sensitivity to steroids and the percentage of Tregs in the bone marrow of children with ALL leukemia is most likely the evidence of the prognostic significance of Tregs for the prognosis of cure. However, if the thesis above would be too daring, then undoubtedly Tregs have at least a significant influence on the response to administered steroids.
It is therefore reasonable to assume that by interfering with immune regulatory system, there might be a possibility to influence the effectiveness of the therapy used, which would consequently lead to the reduction of the dose of therapy without affecting the final therapeutic answer [
At the time of diagnosis, Treg level in the bone marrow also showed a low negative correlation (
Initially, the most important aim of our study was to confirm the observation of an increased percentage of Treg cells in the bone marrow and/or peripheral blood of patients with acute leukemia made by other researchers [
The tumor microenvironment, especially suppression of tumor-associated antigen-reactive lymphocytes, is an important factor in the development and progression of cancer [
Our study revealed statistically higher percentage of Treg cells in the bone marrow than in peripheral blood in the group of 42 children with acute lymphoblastic leukemia. This fact is a very interesting discovery in the context of described infiltration of regulatory lymphocytes into the neoplasmatic tissue in some types of tumors and hematological malignancies. In case of leukemia, such tissue is bone marrow. A similar relationship in the group of patients diagnosed with other condition than acute lymphoblastic leukemia has not been discovered. Hence, there is our interest in these cells in terms of risk factors and biological features of leukemia cells.
It is not known for sure whether the elevated Tregs in the bone marrow is the response to hematological malignancies or the cause of a developing cancer. It is also unclear whether the percentage of regulatory cells correlates with the recognized prognostic factors in acute lymphoblastic leukemia in children. It was only noted that one study on the murine models showed a correlation between the progression of cancer and the migration of regulatory cells into the tumor tissue [
Another cause of elevated number of lymphatic regulatory cells in the bone marrow of children with acute leukemia may be their natural tendency to accumulate in this tissue [
The question remains whether the increased percentage of Treg cells in the bone marrow compared to peripheral blood in children with ALL results from the physiological tendency of Tregs to accumulate in the bone marrow or from the direct contact of immunocompetent cells with blasts.
The study led us to accept the thesis that a higher proportion of Tregs in the bone marrow of children with acute leukemia is due to the interaction of leukemic cells with Tregs. It is less likely that the accumulation of Treg cells in the bone marrow in a such high percentage is a physiological phenomenon.
Interestingly, it seems that there are no statistically significant differences in the proportion of Tregs between BM and PB in patients who suffer from tumors other than ALL [
Detailed analysis of Treg subsets showed very interesting features associated with ALL. The bone marrow of children with ALL was infiltrated by a higher percentage of memory Tregs than the peripheral blood. There was no difference in the number of naive Tregs in both peripheral blood and the bone marrow.
Memory Tregs arise after contact with their own antigen [
The site of this transformation is probably the bone marrow as the percentage of memory Tregs was the highest there. Alternatively, memory Tregs are formed at the periphery and traffic to the sites with high expression of their cognate antigen, such as the leukemic bone marrow.
Most likely, Tregs accumulate there to exert the suppressive effect on the proliferating leukemic blasts. Unfortunately, the increasing percentage of Tregs suppresses also the immune system, which tries to fight a developing tumor. This is probably the reason why the increased percentage of Tregs is seen at the periphery at the very advanced stages of ALL (not analyzed here), while it might be favorable in leukemic bone marrow at early stage of the disease.
Hence, some manipulation on Tregs might be considered as a part of the treatment of hematological malignancies [
Regulatory T lymphocytes are group of cells that might play important role in the development of cancerous diseases including acute leukemia in children. Their elevated bone marrow and peripheral blood rate among children diagnosed with ALL might be linked to the development of the disease. Manipulations involving Tregs might represent an interesting therapeutic option and may be used to enhance the effect of antitumor chemotherapy. Larger studies are now warranted to validate these findings and determine their clinical implications.
Regulatory T cells
Bone marrow
Peripheral blood
Hemoglobin
Acute lymphoblastic leukemia
Chronic lymphoid leukemia
Non-Hodgkin lymphoma
Acute myeloid leukemia
Rhabdomyosarcoma
Lactate dehydrogenase
Berlin-Frankfurt-Munster study group
Standard risk (group)
Intermediate risk (group)
High risk (group)
Event-free survival
Minimal residual disease
Central nervous system
Cluster differentiation antigen
Forkhead box P3
Stromal cell-derived factor 1
Platelet factor 4.
The statistical analysis (as Excel and Statistica files) used to support the findings of this study are available from the corresponding author upon request.
The authors declare that there is no conflict of interest regarding the publication of this paper.